Muutke küpsiste eelistusi

E-raamat: Knowledge Discovery with Support Vector Machines [Wiley Online]

Teised raamatud teemal:
  • Wiley Online
  • Hind: 141,68 €*
  • * hind, mis tagab piiramatu üheaegsete kasutajate arvuga ligipääsu piiramatuks ajaks
Knowledge Discovery with Support Vector MachinesSuurem pilt
Teised raamatud teemal:
Wiley Online e-raamatudRohkem infot Wiley Online kohta
Raamatu kodulehekülg: https://onlinelibrary.wiley.com/doi/book/10.1002/9780470503065

Wiley Series on Methods and Applications in Data Mining

Daniel T. Larose, Series Editor

Knowledge Discovery with Support Vector Machines

Lutz Hamel

An easy-to-follow

introduction to support vector machines

This book provides an in-depth, easy-to-follow introduction to support vector machines drawing only from minimal, carefully motivated technical and mathematical background material. It begins with a cohesive discussion of machine learning and goes on to cover:

  • Knowledge discovery environments

  • Describing data mathematically

  • Linear decision surfaces and functions

  • Perceptron learning

  • Maximum margin classifiers

  • Support vector machines

  • Elements of statistical learning theory

  • Multi-class classification

  • Regression with support vector machines

  • Novelty detection

Complemented with hands-on exercises, algorithm descriptions, and data sets, Knowledge Discovery with Support Vector Machines is an invaluable textbook for advanced undergraduate and graduate courses. It is also an excellent tutorial on support vector machines for professionals who are pursuing research in machine learning and related areas.

PREFACE xiii
PART I 1
1 WHAT IS KNOWLEDGE DISCOVERY?
3
1.1 Machine Learning
4
1.2 Structure of the Universe X
6
1.3 Inductive Learning
8
1.4 Model Representations
9
Exercises
11
Bibliographic Notes
11
2 KNOWLEDGE DISCOVERY ENVIRONMENTS
13
2.1 Computational Aspects of Knowledge Discovery
13
2.1.1 Data Access
14
2.1.2 Visualization
17
2.1.3 Data Manipulation
20
2.1.4 Model Building and Evaluation
23
2.1.5 Model Deployment
26
2.2 Other Tool Sets
27
Exercises
27
Bibliographic Notes
28
3 DESCRIBING DATA MATHEMATICALLY
31
3.1 From Data Sets to Vector Spaces
31
3.1.1 Vectors
35
3.1.2 Vector Spaces
40
3.2 The Dot Product as a Similarity Score
41
3.3 Lines, Planes, and Hyperplanes
44
Exercises
47
Bibliographic Notes
48
4 LINEAR DECISION SURFACES AND FUNCTIONS
49
4.1 From Data Sets to Decision Functions
49
4.1.1 Linear Decision Surfaces Through the Origin
50
4.1.2 Decision Surfaces with an Offset Term
51
4.2 Simple Learning Algorithm
54
4.3 Discussion
57
Exercises
58
Bibliographic Notes
59
5 PERCEPTRON LEARNING
61
5.1 Perceptron Architecture and Training
62
5.2 Duality
67
5.3 Discussion
70
Exercises
71
Bibliographic Notes
72
6 MAXIMUM-MARGIN CLASSIFIERS
73
6.1 Optimization Problems
74
6.2 Maximum Margins
75
6.3 Optimizing the Margin
77
6.4 Quadratic Programming
82
6.5 Discussion
86
Exercises
87
Bibliographic Notes
88
PART II 89
7 SUPPORT VECTOR MACHINES
91
7.1 The Lagrangian Dual
92
7.2 Dual Maximum-Margin Optimization
97
7.2.1 The Dual Decision Function
101
7.3 Linear Support Vector Machines
102
7.4 Nonlinear Support Vector Machines
103
7.4.1 The Kernel Trick
106
7.4.2 Feature Search
109
7.4.3 A Closer Look at Kernels
109
7.5 Soft-Margin Classifiers
114
7.5.1 The Dual Setting for Soft-Margin Classifiers
118
7.6 Tool Support
122
7.6.1 WEKA
123
7.6.2 R
126
7.7 Discussion
128
Exercises
130
Bibliographic Notes
131
8 IMPLEMENTATION
133
8.1 Gradient Ascent
134
8.1.1 The Kernel-Adatron Algorithm
136
8.2 Quadratic Programming
138
8.2.1 Chunking
139
8.3 Sequential Minimal Optimization
142
8.4 Discussion
144
Exercises
144
Bibliographic Notes
145
9 EVALUATING WHAT HAS BEEN LEARNED
147
9.1 Performance Metrics
148
9.1.1 The Confusion Matrix
149
9.2 Model Evaluation
152
9.2.1 The Hold-Out Method
155
9.2.2 The Leave-One-Out Method
157
9.2.3 N-Fold Cross-Validation
158
9.3 Error Confidence Intervals
160
9.3.1 Comparison of Models
162
9.4 Model Evaluation in Practice
163
9.4.1 WEKA
163
9.4.2 R
167
Exercises
169
Bibliographic Notes
170
10 ELEMENTS OF STATISTICAL LEARNING THEORY
171
10.1 The VC-Dimension and Model Complexity
172
10.2 A Theoretical Setting for Machine Learning
175
10.3 Empirical Risk Minimization
176
10.4 VC-Confidence
177
10.5 Structural Risk Minimization
179
10.6 Discussion
180
Exercises
180
Bibliographic Notes
181
PART III 183
11 MULTICLASS CLASSIFICATION
185
11.1 One-Versus-the-Rest Classification
185
11.2 Pairwise Classification
189
11.3 Discussion
192
Exercises
192
Bibliographic Notes
192
12 REGRESSION WITH SUPPORT VECTOR MACHINES
193
12.1 Regression as Machine Learning
193
12.2 Simple and Multiple Linear Regression
194
12.3 Regression with Maximum-Margin Machines
197
12.4 Regression with Support Vector Machines
200
12.5 Model Evaluation
202
12.6 Tool Support
203
12.6.1 WEKA
204
12.6.2 R
205
Exercises
207
Bibliographic Notes
208
13 NOVELTY DETECTION
209
13.1 Maximum-Margin Machines
210
13.2 The Dual Setting
212
13.3 Novelty Detection in R
214
Exercises
217
Bibliographic Notes
217
APPENDIX A NOTATION 219
APPENDIX B TUTORIAL INTRODUCTION TO R 221
B.1 Programming Constructs
222
B.2 Data Constructs
224
B.3 Basic Data Analysis
227
Bibliographic Notes
230
REFERENCES 231
INDEX 237
Lutz Hamel, PhD, teaches at the University of Rhode Island, where he founded the machine learning and data mining group. His major research interests are computational logic, machine learning, evolutionary computation, data mining, bioinformatics, and computational structures in art and literature.
Püsilink: https://www.kriso.ee/db/9780470503065_pe.html
Märksõnad: